Temporary Diabetes Insipidus in 2 Men after On-Pump Coronary Artery Bypass Grafting

Abstract

Many complications have been reported after cardiopulmonary bypass. A common physiologic change during the early postoperative period after cardiopulmonary bypass is increased diuresis. In patients whose urine output is increased, postoperative diabetes insipidus can develop, although reports of this are rare. We present the cases of 2 patients who underwent on-pump coronary artery bypass grafting (with cardiopulmonary bypass). Each was diagnosed with diabetes insipidus postoperatively: a 54-year-old man on the 3rd day, and a 66-year-old man on the 4th day. Each patient recovered from the condition after 6 hours of intranasal therapy with synthetic vasopressin (antidiuretic hormone). The diagnosis of diabetes insipidus should be considered in patients who produce excessive urine early after cardiac surgery in which cardiopulmonary bypass has been used.

Endocrine changes can occur as a response to surgical stress in patients who have undergone cardiac operations with the use of cardiopulmonary bypass (CPB).^1,2 There are few reports of patients in whom diabetes insipidus (DI) has developed after coronary artery bypass grafting (CABG) with the use of CBP (on-pump CABG). The underlying physiopathologic factors of this development have not been clearly defined. We discuss the cases of 2 patients in whom DI developed after on-pump CABG.

Patient 1

In February 2010, 2 days after undergoing a right carotid endarterectomy, a 54-year-old man underwent 3-vessel on-pump CABG. Cardiac protection was provided by means of cold-blood cardioplegic solution. The CPB time was 67 min and the cross-clamp time was 42 min. The patient's minimum intraoperative temperature was 28°C. Only 1,000 mL of crystalloid fluid was given to the patient intraoperatively after CPB was stopped. He remained hemodynamically stable and normotensive. In the intensive care unit, he was given a 5% dextrose solution at a rate of 100 mL/hr for 12 hours. His urine output on the first postoperative day (POD) was 2,550 mL (Table I). No endocrine or neurologic anomalies were noted, and all his laboratory values—including serum electrolyte, creatinine, and urea levels—were within normal ranges. No diuretic therapy was administered.

TABLE I. Daily Urine Volumes of the Patients

Because of the patient's urine output of 12,600 mL on the 2nd POD, fluid replacement was started. The patient was observed to be fatigued and sleepy. His plasma osmolality was 309 mOsm/kg and his urine osmolality was 95 mOsm/kg. In addition, his serum antidiuretic hormone (ADH) level was 0.6 μg/dL (Table II). Computed tomography, substituted for magnetic resonance imaging (MRI) because of the patient's inability to comply, yielded normal results. The patient was diagnosed with DI on the 3rd POD and was started on intranasal desmopressin therapy.

TABLE II. Laboratory Data on the Patients

Patient 2

In April 2011, a 66-year-old man with coronary artery disease underwent 4-vessel on-pump CABG. Cardiac protection was provided by means of cold-blood cardioplegic solution. The CPB time was 112 min and the cross-clamp time was 71 min. The patient's minimum body temperature during the operation was 28°C. Only 1,000 mL of crystalloid fluid was given to the patient intraoperatively after CPB was stopped. He remained hemodynamically stable and normotensive. In the intensive care unit, the patient was given a 5% dextrose solution at a rate of 100 mL/hr for 14 hours. On the first POD, his urine output was 2,200 mL (Table I). All other vital findings and laboratory values on that day were within normal ranges. No diuretic therapy was administered. Because of the patient's urine output of 10,450 mL on the 2nd day and 6,700 mL on the 3rd day, fluid replacement was started. The patient was observed to be fatigued and sleepy. Hyperosmolality in his plasma and hypoosmolality in his urine were detected, and his serum ADH levels were considered to be below normal (Table II). Computed tomography, used because the patient's permanent pacemaker precluded MRI, yielded normal results. The patient was diagnosed with DI on the 4th POD and was started on intranasal desmopressin therapy.

Therapeutic Regimen

During the 6th hour of therapy with desmopressin (synthetic vasopressin), each patient's urine output decreased and urine osmolality began returning to normal levels. The next day, clinical and laboratory improvement was observed: plasma osmolality, urine osmolality, and serum ADH levels returned to normal physiologic levels. The patients' hemodynamic values were within normal ranges. Perioperatively, neither had hypotension, tachycardia, or any cardiac problem. No medication that could have caused DI had been given to either patient perioperatively. Both patients were given oral propranolol on the first POD.

No more desmopressin treatments were required. Patient 1 was discharged from the hospital on the 8th POD and patient 2 on the 10th POD. As of 22 and 8 months later, respectively, each was doing well.

Discussion

Central DI is frequently observed after intracranial surgery. This illness—characterized by lack of urine concentration—yields polyuria despite serum hyperosmolality and decreased intravascular volume caused by ADH deficiency.² However, this disease has been reported only rarely after CPB.

Cardiopulmonary bypass can lead to embolic events that cause neuroendocrine changes and to physiologic changes that cause deterioration in fluid-electrolyte and acid-base equilibria. Although changes in ADH levels have been reported after CPB, the literature about DI after CPB is sparse.^3,4 Three patients of Kuan and colleagues³ developed DI early after CPB and were successfully treated. Ashraf and associates⁵ also observed temporary DI in a patient after CPB.

In accordance with previous observations cited in the medical literature, we observed plasma hyperosmolality and urine hypoosmolality in our patients during the early postoperative period. Their plasma ADH levels were low, which eliminates iatrogenic or psychogenic causes. Microemboli can develop in patients' brains during cardiac surgery and cause certain neurologic problems¹; however, we noted no neurologic sequelae in our patients.

During CPB, volume-loading increases myocyte rigidity. As a result, natriuretic peptide levels rise after CPB. The levels usually return to normal on the 3rd or 4th POD. Clinical DI develops in a very small number of patients.⁶

Clinical and laboratory findings can lead to the conclusion that DI in these patients is caused by vasopressin deficiency. However, it is more likely that DI after CPB results from maladjustment in ADH synthesis, rather than from an inflammatory process or a pathologic lesion. The computed tomographic images in previously reported cases and in our patients yielded normal results, suggesting that DI is a functional condition related to ADH secretion.

Normally, the synthesis of ADH is controlled either osmotically or nonosmotically. Osmotic control is provided by osmoreceptor cells (initially described by Verney⁷) on the front of the hypothalamus. In the circumstance of fluid limitation, extracellular fluid osmolality increases and the volume of the osmoreceptor cells decreases. This event triggers ADH secretion. These receptor cells can be sensitive to changes of as little as 1% to 2% in extracellular fluid osmolality. The nonosmotic control of ADH secretion is provided by low-volume receptors in the left atrium. A decrease in left atrial volume leads to ADH secretion from the hypothalamus in parasympathetic fashion via the vagus nerve.⁸ Most patients experience no clinical problems, because their plasma osmolality is adjusted by the osmoreceptors.⁹ However, in other individuals, the osmoreceptor cells barely affect ADH secretion; ADH secretion is nonosmotically controlled, and only by the left atrial cells.¹⁰ In other words, osmoreceptors might sometimes be insensitive to changes in plasma osmolality, independent of whether CPB has been used. When the function of such patients' left atrial nonosmotic receptors temporarily deteriorates because of the use of cardioplegia or for other reasons, ADH secretion also decreases. It could take 10 days for the left atrial cells to return to normal function. After those cells resume control of ADH secretion, external ADH supplementation is no longer necessary.

As many as 1,200 patients undergo on-pump cardiac operations in our clinic every year. Two such patients, both of whom had undergone CABG, were recently diagnosed with temporary DI. Our report serves as a reminder that patients who undergo cardiac surgery with the use of CPB could develop DI because of polyuria during the postoperative period. We found that prompt therapy with vasopressin prevented further problems.